Fraction vegetation cover extraction of winter wheat based on RGB image obtained by UAV

Authors:
  • JZJing Zhao
    Jing Zhao
  • YLYubin Lan
    Yubin Lan
  • LLLiqun Lu
    Liqun Lu
  • ZLZhiming Li
    Zhiming Li

Abstract
Abstract: In order to quickly and accurately extract the fraction vegetation cover (FVC) of winter wheat, the unmanned aerial vehicle (UAV) was used to obtain the visible light image of winter wheat during the green returned stage, and the estimation ability of winter wheat FVC based on visible light vegetation index and texture feature was explored, 8 texture features of winter wheat images were extracted by gray-level co-occurrence matrix, and 3 visible light vegetation indices of VDVI, NGBDI and GRVI were calculated. According to the characteristics of vegetation and soil blue & green and green & red bands in the research field, the green-red combined vegetation index (GRDI) and green-blue combined vegetation index (GBVI) were constructed. In addition, regression models of vegetation cover were established using 5 vegetation indices and 8 texture features respectively. The results showed that the regression model established by GRDI had the highest accuracy among vegetation indices with R2 value of 0.9245, RMSE value of 0.02677, and nRMSE value of 5.74%. Therefore, the GRDI model was selected in this paper to generate FVC level distribution map, which provided a basis for winter wheat growth monitoring and field management. Keywords: winter wheat, FVC, UAV remote sensing technology, texture feature, vegetation index construction DOI: 10.33440/j.ijpaa.20190202.44. Citation: Yang H B, Zhao J, Lan Y B, Lu L Q, Li Z M. Fraction vegetation cover extraction of winter wheat based on spectral information and texture features obtained by UAV. Int J Precis Agric Aviat, 2019; 2(2): 54–61.